Set the purpose and boundaries
Before combining models, the project team should state which areas, systems, levels, and milestones are included. Define whether the review is looking for geometric intersection, clearance, access, route continuity, or another stated condition. A clash check outside these boundaries can create noise, while an undefined boundary can leave important spaces unexamined. Coordination scope should also state what is excluded, such as concealed conditions that are not represented or design questions that require a responsible engineer.
- Name included disciplines, areas, and model revisions
- Set review criteria and tolerance rules
- List known exclusions and unresolved design inputs
Establish reliable shared inputs
A combined review depends on compatible coordinate origins, units, orientation, levels, and model versions. Teams should record each file's source, revision, date received, authoring context, and intended status. Backgrounds that have shifted or outdated architectural and structural references can create false conflicts. Model alignment is a technical check, not proof that the source design is current. When there is uncertainty, stop and resolve it with the model owner before drawing conclusions from the combined view.
Model the constraints that matter
A useful coordination model represents more than equipment outlines. Depending on the agreed scope, it may need insulation, slopes, fittings, supports, access envelopes, service clearances, and the space required to install or replace components. The required information varies by stage and task. Overly coarse geometry can conceal a real issue; unnecessary detail can consume effort and create an impression of certainty that the source information does not support. State the expected level of representation for each system.
- Include access and maintenance envelopes where required
- Check slopes and transitions for relevant systems
- Distinguish modeled geometry from assumed geometry
Use a repeatable review sequence
A practical review begins with file validation and area selection, then checks major routes and constrained zones before examining smaller local conditions. Automated geometric tests can identify candidate intersections using agreed rules. Human review determines whether a result is meaningful, whether the geometry is reliable, and which discipline should respond. Teams should group duplicates, suppress documented intentional relationships, and retain a record of why a finding was closed. A clear review sequence makes successive revisions easier to compare.
Record issues for decisions
An issue record should let another person find and understand the condition without relying on an informal meeting memory. Include location, affected systems, source revisions, a view or reference, the observed condition, and the question requiring a response. Assign a proposed owner and due milestone only through the project's agreed process. Distinguish a potential intersection from a design conflict: the latter requires context and a decision by the responsible discipline. A coordination record identifies a question; it does not prescribe an engineering solution.
- Keep status and revision history visible
- Describe evidence separately from proposed action
- Route design decisions to the appointed discipline lead
Review constructability and practical access
Geometric separation does not prove that systems can be installed, supported, inspected, or maintained. Teams may need to consider installation sequence, lifting paths, access panels, fire stopping, structural openings, support locations, and replacement routes. These checks require the relevant construction and operations participants and project specific information. A model review should flag questions beyond its agreed scope rather than silently treating a visually clear arrangement as buildable or maintainable.
Close the loop and hand off
At each exchange, publish the combined model or issue record with its input list, revision identifiers, review date, coverage, criteria, and open findings. Model authors update their own discipline information and identify changes made in response. A coordinator can assemble and report the review, but approval authority remains with the project roles defined by contract and governance. Before a milestone, the team should decide which issues are accepted, deferred, or still require resolution, and carry those decisions into the next exchange.
Control coordination noise
Automated tests can produce many candidate results that are duplicates, intentional conditions, or below the project's practical tolerance. Agree how these cases are categorized and who may mark them as reviewed. Keep a reason for suppression so a later model revision can be checked rather than inheriting an unexplained closed status. Report coverage alongside issue totals, since a count has little meaning without the areas and rules tested. Revisit thresholds when geometry, construction stage, or project priorities change.